Scrutable

I’ve seen a motorway gantry hit by a lorry that had a raised tipper back. Just saying.

Might forward this thread to some activist friends.

Fossil infrastructure can be very vulnerable - useful reminder for those working to hasten the death of fossil fuels.

tom p wrote: Sat Jan 15, 2022 3:25 pm 58 mph, innit?

60 (by speed limiter - not just by law). It used ot be 56 but was raised in 2015.

And, as others have mentioned, just in time is entirely unnecessary. Half a mile before the obstruction should be a minimum, along with multiple sensors either side of the motorway, like the laser 'trip wires' you see in museums in hollywood movies which would warn any lorry that still had its pantograph up. That gives them 30 seconds to hit the emergency pantograph retraction mechanism.
Couple that with strict liability for the lorry drivers & owners/lessees for all costs associated with a pantograph retraction failure and no f.ckers gonna crash theirs.

Anything requiring drivre intervention won't work. See 11foot8.com for many videos of one bridge in the US. And fancy high-tech stuff goes wrong all the time. It needs a purely mechanical backup, even if there's a high-tech mechanism that usually handles it.

Anything that depends on the haulage industry maintaining its vehicles properly, or complying with the law, won't work.

"Tripwire" style alarms with light beams are all very well but in the end I like the last-ditch practicality of the devices built to prevent bridge strikes which are "sacrificial" barriers built from hefty steel girders a little way short of the bridge, at the same height and painted with bright yellow and black stripes. If you ignore all warnings you will hit that before you get close to touching the bridge itself.

Martin Y wrote: Sun Jan 16, 2022 12:33 pm "Tripwire" style alarms with light beams are all very well but in the end I like the last-ditch practicality of the devices built to prevent bridge strikes which are "sacrificial" barriers built from hefty steel girders a little way short of the bridge, at the same height and painted with bright yellow and black stripes. If you ignore all warnings you will hit that before you get close to touching the bridge itself.

There are lots of those in the hilly places between here and Slovakia where the motorway passes through actual tunnels.

Martin Y wrote: Sun Jan 16, 2022 12:33 pm "Tripwire" style alarms with light beams are all very well but in the end I like the last-ditch practicality of the devices built to prevent bridge strikes which are "sacrificial" barriers built from hefty steel girders a little way short of the bridge, at the same height and painted with bright yellow and black stripes. If you ignore all warnings you will hit that before you get close to touching the bridge itself.

Like this one?

Yes. If only I could find the photo I have one from the bridge by Langley station near Slough, with the shaved-off roof of a double decker bus on the road below it. It was a school bus, but thankfully returning empty. Driver was pissed.

Approaching the Port at Le Havre, there is a 'comb' that would hit too-tall vehicles that would be at risk of running into the raised walkway further down the road

Martin Y wrote: Sun Jan 16, 2022 12:55 pm Yes. If only I could find the photo I have one from the bridge by Langley station near Slough, with the shaved-off roof of a double decker bus on the road below it. It was a school bus, but thankfully returning empty. Driver was pissed.

This one?

stańczyk wrote: Sun Jan 16, 2022 1:28 pm This one?

Why, yes. Thanks. The very same.

Basic self drive technology, already available in say a Tesla, tells the car the speed limit and that there's a roundabout coming up and slows down by itself accordingly. There's the first stage of car to car communication underway to warn of sudden deceleration on the motorway ahead. Satnavs no longer just communicate to the driver, they communicate to the car directly.

If you're getting a new fleet of electric buses it's the easiest thing in the world to get automated prevention of going under low bridges. I've got a suspicion people don't know how advanced self-drive has become in the last couple of years, because the full self-drive technology of zero humans now seems further away. But chuck in a bunch of sensors and cameras and the latest satnavs and vehicles simply won't be able to crash into low bridges.

lpm wrote: Sun Jan 16, 2022 3:35 pm Basic self drive technology, already available in say a Tesla, tells the car the speed limit and that there's a roundabout coming up and slows down by itself accordingly. There's the first stage of car to car communication underway to warn of sudden deceleration on the motorway ahead. Satnavs no longer just communicate to the driver, they communicate to the car directly.

If you're getting a new fleet of electric buses it's the easiest thing in the world to get automated prevention of going under low bridges. I've got a suspicion people don't know how advanced self-drive has become in the last couple of years, because the full self-drive technology of zero humans now seems further away. But chuck in a bunch of sensors and cameras and the latest satnavs and vehicles simply won't be able to crash into low bridges.

And you even need anything as advanced as "self-driving" technology that has to built specifically into vehicles for that, with cameras and computer vision and sh.t. GPS alone can do it.

dyqik wrote: Sun Jan 16, 2022 9:27 pm

lpm wrote: Sun Jan 16, 2022 3:35 pm Basic self drive technology, already available in say a Tesla, tells the car the speed limit and that there's a roundabout coming up and slows down by itself accordingly. There's the first stage of car to car communication underway to warn of sudden deceleration on the motorway ahead. Satnavs no longer just communicate to the driver, they communicate to the car directly.

If you're getting a new fleet of electric buses it's the easiest thing in the world to get automated prevention of going under low bridges. I've got a suspicion people don't know how advanced self-drive has become in the last couple of years, because the full self-drive technology of zero humans now seems further away. But chuck in a bunch of sensors and cameras and the latest satnavs and vehicles simply won't be able to crash into low bridges.

And you even need anything as advanced as "self-driving" technology that has to built specifically into vehicles for that, with cameras and computer vision and sh.t. GPS alone can do it.

GPS is only as good as the maps being used. Cameras and
sensors will help to prevent crashing into anything not on the map (eg fallen trees or new construction).

Woodchopper wrote: Sun Jan 16, 2022 10:07 pm

dyqik wrote: Sun Jan 16, 2022 9:27 pm

lpm wrote: Sun Jan 16, 2022 3:35 pm Basic self drive technology, already available in say a Tesla, tells the car the speed limit and that there's a roundabout coming up and slows down by itself accordingly. There's the first stage of car to car communication underway to warn of sudden deceleration on the motorway ahead. Satnavs no longer just communicate to the driver, they communicate to the car directly.

If you're getting a new fleet of electric buses it's the easiest thing in the world to get automated prevention of going under low bridges. I've got a suspicion people don't know how advanced self-drive has become in the last couple of years, because the full self-drive technology of zero humans now seems further away. But chuck in a bunch of sensors and cameras and the latest satnavs and vehicles simply won't be able to crash into low bridges.

And you even need anything as advanced as "self-driving" technology that has to built specifically into vehicles for that, with cameras and computer vision and sh.t. GPS alone can do it.

GPS is only as good as the maps being used. Cameras and
sensors will help to prevent crashing into anything not on the map (eg fallen trees or new construction).

GPS maps for major highways in the western world are good to within a meter or two. Easily good enough for this application. The maps are significantly better than the actual live positions.

I find the speed limit changes in Google maps are good to within 5 seconds when driving. Which is about as often as I can look at the screen while driving. Which is easily good enough to give a minutes notice to lower a pantograph (which would take ~30s), with some significant margin.

In any case, the damage done by a pantograph hitting a bridge is going to be irrelevant to the bridge's structural integrity.

Someone will have to go and pick the debris off the road, and the truck will need to be fixed. And that's about it.

dyqik wrote: Sun Jan 16, 2022 10:50 pm In any case, the damage done by a pantograph hitting a bridge is going to be irrelevant to the bridge's structural integrity.

Someone will have to go and pick the debris off the road, and the truck will need to be fixed. And that's about it.

There are loads of overhead structures on motorways, that are presumably at the same height as bridges. Especially smart motorways. And bridges covers a lot of different structures, some of which are only for pedestrians.

Grumble wrote: Sun Jan 16, 2022 11:32 pm

dyqik wrote: Sun Jan 16, 2022 10:50 pm In any case, the damage done by a pantograph hitting a bridge is going to be irrelevant to the bridge's structural integrity.

Someone will have to go and pick the debris off the road, and the truck will need to be fixed. And that's about it.

There are loads of overhead structures on motorways, that are presumably at the same height as bridges. Especially smart motorways. And bridges covers a lot of different structures, some of which are only for pedestrians.

All of which means that bridge strikes by pantographs would be vanishingly rare. Because operators who didn't make the system work would lose all their equipment really quickly.

lpm wrote: Sun Jan 16, 2022 3:35 pm Basic self drive technology, already available in say a Tesla, tells the car the speed limit and that there's a roundabout coming up and slows down by itself accordingly.

Maybe, but have they fixed the problem of driving into stationary objects yet?

dyqik wrote: Sun Jan 16, 2022 10:41 pm

Woodchopper wrote: Sun Jan 16, 2022 10:07 pm

dyqik wrote: Sun Jan 16, 2022 9:27 pm

And you even need anything as advanced as "self-driving" technology that has to built specifically into vehicles for that, with cameras and computer vision and sh.t. GPS alone can do it.

GPS is only as good as the maps being used. Cameras and
sensors will help to prevent crashing into anything not on the map (eg fallen trees or new construction).

GPS maps for major highways in the western world are good to within a meter or two. Easily good enough for this application. The maps are significantly better than the actual live positions.

I find the speed limit changes in Google maps are good to within 5 seconds when driving. Which is about as often as I can look at the screen while driving. Which is easily good enough to give a minutes notice to lower a pantograph (which would take ~30s), with some significant margin.

The issue isn’t the per-meter accuracy of the maps. It’s whether the maps have been updated with every obstacle that could be hit by a pantograph. Cameras and other sensors are a solution to human error.

Hmmm, how could you possibly get information about major roadworks on busy highways that take years to plan and approve to mapping systems in time.

dyqik wrote: Mon Jan 17, 2022 2:50 am Hmmm, how could you possibly get information about major roadworks on busy highways that take years to plan and approve to mapping systems in time.

You would, of course, get people involved to update the information into databases. Which is the problem as every human system is prone to error.

It’s easy to think of examples of how errors are made. The person entering the data made a mistake. Someone responsible for deciding what data should be entered misunderstood the situation. Someone else suddenly went on long term sick leave and the data never got entered. People working on the roads make an operational decision to change plans but don’t update the database. We could spend the rest of the day thinking up other examples.

Perhaps it could be possible to build an error free human system. But it’ll be far easier to equip the vehicles with sensors.

These problems aren’t a major problem for road safety at the moment because every vehicle has a human driver who looks at the environment around them and override errors in the digital maps. Which is what I needed to do when driving in the UK over Christmas.

On the point of electrification, a few things to mention. Electrifying underneath overbridges is one of the areas I've worked on in my current job, producing a tool to help project teams make decisions about what to do. There are three main options:

1. Raise the bridge. Expensive, and closes both the railway and the road (if the bridge carries a road) for a time. Problematic if the bridge is listed.

2. Lower the track. Doable but annoying and expensive. Shuts the railway for ages and can introduce drainage issues and the risk of wet beds which affect track condition. Required to be done over a fair run up each side of the bridge.

3. Accept reduced clearance between the contact wire and the bridge deck. The easiest of the option but introduces reliability problems and a higher risk of flashovers which shortens the life of the system, can dewire the wire run or destroy the power supply and cause millions of pounds of damage - all it takes is rain and the wrong pigeon in the wrong place at the wrong time. This option may not be possible if there is a station or level crossing near the bridge, as lower clearance may not be legal.

In all my chats with the engineers, they've never mentioned EU regs as a preventative for selecting option 3. Instead, they've treated it as an engineering problem requiring only the right solution. Options include contact rail, insulating paint covering the top of the system to prevent arcing, and (as mentioned here) neutral sections under the span. That last option is how the Paisley Canal branch line was electrified so quickly and cheaply. It does leave the trains at risk of being stranded, especially if the signals are in the wrong place. It's not really seen though as a workable option for main lines, but could be used more for more rural areas without too much problem.

Even more solutions are being looked at so that options 1 and 2 are rarely or never needed.

Martin Y wrote: Sun Jan 16, 2022 12:33 pm "Tripwire" style alarms with light beams are all very well but in the end I like the last-ditch practicality of the devices built to prevent bridge strikes which are "sacrificial" barriers built from hefty steel girders a little way short of the bridge, at the same height and painted with bright yellow and black stripes. If you ignore all warnings you will hit that before you get close to touching the bridge itself.

Yeah, those are great too.
Belt AND braces. Over-engineer it and have multiple failsafes both leading up to the bridges & on the lorries, knowing full well that there will be some sh.tty lorry companies & some dozy drivers and you may well get more than one in a day before the first strike has been repaired

Millennie Al wrote: Mon Jan 17, 2022 2:06 am

lpm wrote: Sun Jan 16, 2022 3:35 pm Basic self drive technology, already available in say a Tesla, tells the car the speed limit and that there's a roundabout coming up and slows down by itself accordingly.

Maybe, but have they fixed the problem of driving into stationary objects yet?

I think this comment proves my point that "I've got a suspicion people don't know how advanced self-drive has become in the last couple of years, because the full self-drive technology of zero humans now seems further away."

These cars almost always detect stationary objects and don't drive into them. They need a human as a backup sensor for the rare occasions when they fail to detect them. Cars with the necessary perfection to do without the human seem a distant dream at this point.